Building complex hierarchical model construction method, device, equipment, medium and program product

Abstract

The present application relates to a method, device, equipment, medium and program product for constructing a building complex hierarchical model. The method comprises: performing plane detection on the original model data of the original building complex model through a regional growth model, and performing voxel grouping on the obtained initial plane data, and then performing plane aggregation to obtain aggregated structural segments; optimizing the model according to the aggregated structural segments and a preset initial energy model to achieve the main hierarchical division of the aggregated structural segments to obtain the main building segments and the outer building segments; then performing outer layer division on the outer building segments to obtain the outer building layer data; finally, constructing a triangular mesh model according to the outer building layer data and the main building segments to obtain the building complex hierarchical model. The present method can ensure the consistency of the model structure within the same LOD level of the building complex hierarchical model without the need for pre-defining the levels.

Description

Technical Field

[0001] The present application relates to the field of computer graphics technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium and computer program product for constructing a building complex hierarchical model. Background Art

[0002] With the development of computer graphics technology, level of detail model (LOD) generation technology has emerged. LOD generation technology can significantly improve rendering efficiency and reduce the consumption of computing resources by effectively managing hierarchical details, while maintaining visual detail and realism. In traditional technology, LOD generation technology for architectural models is currently divided into two categories: one is a method based on predefined levels, which lacks flexibility when dealing with models with complex geometric structures. The other is to generate LOD based on the geometry or visual information of the model itself. This method is difficult to maintain the structural characteristics of the original input when generating LOD models with extremely low complexity. Therefore, there is an urgent need for a method for constructing a hierarchical model of a building complex that can ensure the consistency of the model structure within the same LOD level without the need for predefined levels. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for constructing a hierarchical model of a building complex, which can ensure the consistency of the model structure within the same LOD level without the need for pre-defined levels, in order to address the above technical problems.

[0004] In a first aspect, the present application provides a method for constructing a building complex hierarchical model, comprising:

[0005] Performing plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model;

[0006] voxelize and group the initial plane data to obtain a voxel plane group, and perform plane aggregation on the voxel plane group to obtain an aggregated structure segment;

[0007] Performing model optimization according to the aggregate structure segmentation and the preset initial energy model to obtain a target energy model;

[0008] According to the target energy model, the aggregate structure segment is divided into main hierarchical groups to obtain building main segments and building outer segments;

[0009] Dividing the outer layer of the building into segments into outer layers to obtain outer layer data of the building;

[0010] A triangular mesh model is constructed according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0011] In one embodiment, the voxelizing and grouping the initial plane data to obtain a voxel plane group includes:

[0012] voxelize the original model data to obtain voxel units;

[0013] The initial plane data is plane-grouped according to the voxel units to obtain a voxel plane group.

[0014] In one embodiment, the performing planar aggregation on the voxel plane group to obtain aggregated structure segments comprises:

[0015] performing overlap evaluation on the voxel plane group and evaluating the result;

[0016] The voxel plane groups are aggregated according to the evaluation results to obtain aggregated structure segments.

[0017] In one embodiment, the step of performing model optimization according to the aggregate structure segmentation and the preset initial energy model to obtain the target energy model includes:

[0018] Performing binary linear programming on the aggregation structure segment by segment to obtain linear programming results;

[0019] The preset initial energy model is solved by linear programming according to the linear programming result to obtain the target energy model.

[0020] In one embodiment, the step of dividing the outer layer of the building into segments to obtain the outer layer data of the building includes:

[0021] Performing spectral clustering analysis on the outer layer segments of the building to obtain outer layer level clustering results;

[0022] The outer layer segments of the building are hierarchically divided according to the outer layer clustering result to obtain the outer layer data of the building.

[0023] In one embodiment, the preset initial energy model includes an initial simplicity sub-model, an initial semantic sub-model and an initial shape fidelity sub-model. The method further includes:

[0024] A shape-fidelity model is constructed by using the intersection of the obtained initial associated structure segment set and the obtained initial main structure segment set to obtain an initial shape-fidelity sub-model;

[0025] A simplicity model is constructed by using the plane area ratio between the initial associated structure segment set and the initial main structure segment set to obtain an initial simplicity sub-model;

[0026] Constructing a semantic model by cross-similarity between the initial associated structure segment set and the initial main structure segment set to obtain an initial semantic sub-model;

[0027] An energy model is constructed according to the shape-fidelity sub-model, the initial simplicity sub-model and the initial semantic sub-model to obtain the initial energy model.

[0028] In a second aspect, the present application also provides a device for constructing a building complex hierarchical model, comprising:

[0029] A plane detection module is used to perform plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model;

[0030] A structure segmentation generation module, used for performing voxel grouping on the initial plane data to obtain a voxel plane group, and performing plane aggregation on the voxel plane group to obtain an aggregated structure segment;

[0031] A model optimization module, used to optimize the preset initial energy model according to the aggregation structure segmentation to obtain a target energy model;

[0032] A main body division module, used for performing main body hierarchical division on the aggregate structure segment according to the target energy model to obtain a building main body segment and a building outer layer segment;

[0033] An outer layer division module is used to divide the outer layer segments of the building into outer layers to obtain outer layer data of the building;

[0034] The model building module is used to construct a triangular mesh model according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0035] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0036] Performing plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model;

[0037] voxelize and group the initial plane data to obtain a voxel plane group, and perform plane aggregation on the voxel plane group to obtain an aggregated structure segment;

[0038] Performing model optimization according to the aggregate structure segmentation and the preset initial energy model to obtain a target energy model;

[0039] According to the target energy model, the aggregate structure segment is divided into main hierarchical groups to obtain building main segments and building outer segments;

[0040] Dividing the outer layer of the building into segments into outer layers to obtain outer layer data of the building;

[0041] A triangular mesh model is constructed according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0042] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0043] Performing plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model;

[0044] voxelize and group the initial plane data to obtain a voxel plane group, and perform plane aggregation on the voxel plane group to obtain an aggregated structure segment;

[0045] Performing model optimization according to the aggregate structure segmentation and the preset initial energy model to obtain a target energy model;

[0046] According to the target energy model, the aggregate structure segment is divided into main hierarchical groups to obtain building main segments and building outer segments;

[0047] Dividing the outer layer of the building into segments into outer layers to obtain outer layer data of the building;

[0048] A triangular mesh model is constructed according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0049] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0050] Performing plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model;

[0051] voxelize and group the initial plane data to obtain a voxel plane group, and perform plane aggregation on the voxel plane group to obtain an aggregated structure segment;

[0052] Performing model optimization according to the aggregate structure segmentation and the preset initial energy model to obtain a target energy model;

[0053] According to the target energy model, the aggregate structure segment is divided into main hierarchical groups to obtain building main segments and building outer segments;

[0054] Dividing the outer layer of the building into segments into outer layers to obtain outer layer data of the building;

[0055] A triangular mesh model is constructed according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0056] The above-mentioned building complex hierarchical model construction method, device, computer equipment, computer-readable storage medium and computer program product perform plane detection on the original model data through the regional growth model to obtain initial plane data, wherein the original model data is used to store the original building complex model, and then the initial plane data is voxelized and grouped to obtain voxel plane groups, and the voxel plane groups are plane aggregated to obtain aggregated structural segments, so that each building in the original building complex model is divided into multiple aggregated structural segments, which is more flexible than the pre-defined hierarchy method and also has better structural perception ability. Subsequently, the model is optimized according to the aggregated structural segments and the preset initial energy model to obtain a target energy model that can better perceive the detailed structure of the aggregated structural segments, and then the aggregated structural segments are divided into the main hierarchical level according to the target energy model, and the aggregated structural segments are divided into the main building segments and the outer building segments. The hierarchical division through the optimized target energy model can ensure the structural consistency across buildings in the same level. Then, the outer layer of the building is divided into outer layers to obtain the outer layer data of the building. According to the outer layer data of the building and the main body of the building, a triangular mesh model is constructed to obtain a building complex hierarchical model. The building complex hierarchical model is a high-resolution original building complex model, which realizes the step-by-step fine representation from the low-resolution original building complex model to the high-resolution building complex hierarchical model. Therefore, this solution realizes a building complex hierarchical model construction method that can ensure the consistency of the model structure within the same LOD level without pre-defining the level. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A diagram of an application environment of a method for constructing a building complex hierarchical model in one embodiment;

[0059] Figure 2 A schematic diagram of a process of constructing a building complex hierarchical model in one embodiment;

[0060] Figure 3 A schematic diagram of a process of constructing a building complex hierarchical model in one embodiment;

[0061] Figure 4 A schematic diagram of a flow chart of a method for constructing a building complex hierarchical model in another embodiment;

[0062] Figure 5 It is a structural block diagram of a device for constructing a building complex hierarchical model in one embodiment;

[0063] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] The building complex hierarchical model construction method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 performs plane detection on the original model data through the regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model; the initial plane data is voxelized and grouped to obtain voxel plane groups, and the voxel plane groups are plane aggregated to obtain aggregated structure segments; the model is optimized according to the aggregated structure segments and the preset initial energy model to obtain the target energy model; according to the target energy model, the aggregated structure segments are divided into main hierarchical levels to obtain building main segments and building outer segments; the building outer segments are divided into outer layers to obtain building outer layer data; a triangular mesh model is constructed according to the building outer layer data and the building main segments to obtain a building complex hierarchical model; the building complex hierarchical model is a high-resolution original building complex model. The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0066] In an exemplary embodiment, Figure 2 As shown, a method for constructing a building complex hierarchical model is provided, and the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 202 to 212. Among them:

[0067] Step 202, performing plane detection on the original model data through the regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model.

[0068] The regional growth model may adopt a Region Growing algorithm model, and the original building complex model may be a building complex model including multiple building models.

[0069] It should be noted that the building model is usually stored in the form of a point cloud set or a triangular mesh. The original model data can be point cloud data corresponding to the point cloud set storage form, or can be mesh data in the triangular mesh storage form, but is not limited thereto.

[0070] In some embodiments, the original model data is subjected to initial plane detection by a region growing algorithm model, and the α shape data of the original model data is calculated to define the boundary of each initial plane, thereby obtaining the initial plane data, wherein the α shape data is α-shape.

[0071] Step 204 , voxelize and group the initial plane data to obtain voxel plane groups, and perform plane aggregation on the voxel plane groups to obtain aggregated structure segments.

[0072] The voxel grouping is a voxel grouping process based on the bounding box of the original building complex model, and the plane aggregation process is a plane aggregation process based on the overlapping ratio between voxel plane groups, wherein the bounding box can be the Bbox of the original building complex model.

[0073] In some embodiments, voxelizing and grouping the initial plane data to obtain a voxel plane group includes: voxelizing the original model data to obtain voxel units; and plane grouping the initial plane data according to the voxel units to obtain a voxel plane group.

[0074] In some embodiments, the original model data is voxelized based on the bounding box of the original building complex model to obtain n*n*n discrete voxel units for subsequent plane grouping processing.

[0075] The voxelization process may be uniform voxelization or adaptive spatial discretization to improve computational efficiency, but is not limited thereto.

[0076] In some embodiments, the initial plane data is plane-grouped according to voxel units, and a voxel plane group surrounding each voxel can be obtained by emitting rays outward from each voxel unit for use in a subsequent plane aggregation step.

[0077] In some embodiments, performing planar aggregation on the voxel plane group to obtain the aggregated structure segment includes: performing overlap evaluation on the voxel plane group and obtaining an evaluation result; and aggregating the voxel plane group according to the evaluation result to obtain the aggregated structure segment.

[0078] In some embodiments, the voxel plane groups corresponding to adjacent voxel units are evaluated one by one. If the overlap rate of the two voxel plane groups is greater than 50%, the evaluation result is that there is a large overlap between the two voxel plane groups. If the overlap rate of the two voxel plane groups is less than or equal to 50%, the evaluation result is that there is a small overlap between the two voxel plane groups.

[0079] In some embodiments, based on the evaluation results, two voxel plane groups having more overlapping parts are merged to obtain aggregated structure segments, and two voxel units corresponding to the two voxel plane groups are also merged into a new voxel unit.

[0080] Step 206 , performing model optimization according to the aggregate structure segmentation and the preset initial energy model to obtain a target energy model.

[0081] The expression of the preset initial energy model is as follows:

[0082]

[0083] in, represents the initial shape-preserving sub-model, represents the initial simplicity submodel and represents the initial semantic sub-model, n represents the number of buildings in the original building complex model, Representative and original building complex model A collection of aggregated structural segments associated with a building, Representative identification The aggregated structural segment collection of the main building segment, and There are two parameters used to adjust and Impact on the initial energy model. represents the energy fraction of the initial energy model, A higher value means that the main building segment structure corresponding to the current initial energy model is better.

[0084] In some embodiments, model optimization is performed based on aggregate structure segments and a preset initial energy model to obtain a target energy model, including: performing binary linear programming on the aggregate structure segments to obtain linear programming results; performing linear programming on the preset initial energy model based on the linear programming results to obtain the target energy model.

[0085] In some embodiments, by dividing each polymer structure segment into a building main body segment or a building outer layer segment, binary linear programming is performed on the polymer structure segment based on the above division method to obtain a linear programming result that can represent all division situations.

[0086] Among them, in the binary linear programming, if the aggregate structure segment is divided into the building main body segment, it is represented as 1, and if it is divided into the building outer layer segment, it is represented as 0.

[0087] In some embodiments, each linear programming result is brought into a preset initial energy model to calculate the energy score. When the energy score is the highest, the energy model corresponding to the energy score is used as the target energy model. At this time, the linear programming result of each aggregate structure segment corresponding to the target energy model is the linear programming optimal solution of the preset initial energy model.

[0088] Step 208 , dividing the aggregate structure segments into main layers according to the target energy model to obtain building main segments and building outer layers.

[0089] In some embodiments, the aggregate structure segments are divided into main hierarchical groups according to the linear programming results corresponding to the target energy model. The aggregate structure segments marked as 1 in the linear programming results are divided into building main segments, and the aggregate structure segments marked as 0 in the linear programming results are divided into building outer segments. Finally, all the aggregate structure segments are divided to obtain building main segments and building outer segments.

[0090] Step 210, dividing the building outer layer segments into outer layer levels to obtain building outer layer level data.

[0091] Among them, the outer layer division processing is used to divide the outer layer of the building into more detailed hierarchical segments, so as to further improve the precision of the subsequent construction of the building complex hierarchical model.

[0092] In some embodiments, outer-layer segmentation of the building is performed to obtain outer-layer data of the building, including: performing spectral clustering analysis on the outer-layer segmentation of the building to obtain outer-layer clustering results; and hierarchically dividing the outer-layer segmentation of the building according to the outer-layer clustering results to obtain outer-layer data of the building.

[0093] In some embodiments, the building exterior segments are assigned to different building exterior levels based on the cross-similarity between the building exterior segments. The similarity matrix and the normalized Laplacian matrix of the cross-similarity between the building exterior segments are as follows:

[0094]

[0095] in, is a similarity matrix constructed by the cross-similarity between building exterior segments, [i,j] is the similarity between structure segments i and j, Normalized Laplacian matrix for the building envelope segments.

[0096] In some embodiments, by The eigenvectors corresponding to the first X smallest eigenvalues ​​are used as the segmentation features of the building exterior segment, and the k-means method is used to cluster the segmentation features, so that the building exterior segment is clustered into X different building exterior levels, and the building exterior level to which each building exterior segment belongs is recorded in the exterior level clustering result.

[0097] In some embodiments, the building exterior segments are allocated to corresponding levels according to the number of levels recorded in the exterior level clustering results and the building exterior level to which each building exterior segment belongs, thereby achieving level division and obtaining building exterior level data that can record all building exterior segments in each level.

[0098] Step 212, construct a triangular mesh model based on the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0099] In some embodiments, a binary space partitioning technique is used, and the bounding box of the building model is partitioned by planes based on the building outer layer data and the building main body segment. LOD0 is partitioned using the planes contained in the building main body segment belonging to LOD0, and LODX in the building outer layer data is partitioned using the planes contained in the structure segments of LOD1...LODX to obtain the plane data contained in each level. Among them, LODX can be the X-level data in the building outer layer data, and the binary space partitioning technique can be the BSP technique. Based on the convex hull corresponding to each plane data, the light rays are simulated, and the internal and external attributes of each convex hull are determined by the shooting statistics of the rays inside or outside the convex hull. If the angle between the light emitted from the inside of the convex hull and the building is less than 90 degrees, it is marked as an internal convex hull, and then all the convex hulls marked as the inside are extracted from the triangular mesh model to obtain the surface data of the building complex hierarchical model, and the plane data contained in each level and the surface data of the building complex hierarchical model are constructed into a triangular mesh model to obtain the building complex hierarchical model.

[0100] In the above-mentioned building complex hierarchical model construction method, the original model data is plane-detected by the regional growth model to obtain the initial plane data, wherein the original model data is used to store the original building complex model, and then the initial plane data is voxelized and grouped to obtain the voxel plane group, and the voxel plane group is plane-aggregated to obtain the aggregated structure segment, so that each building in the original building complex model is divided into multiple aggregated structure segments, which is more flexible than the pre-defined hierarchy method and also has better structural perception ability. Subsequently, the model is optimized according to the aggregated structure segment and the preset initial energy model to obtain the target energy model that can better perceive the detailed structure of the aggregated structure segment, and then the aggregated structure segment is divided into the main building segment and the outer building segment according to the target energy model. The hierarchical division through the optimized target energy model can ensure the structural consistency across buildings in the same level. Then the outer layer of the building is divided into outer layers to obtain the outer layer data of the building. According to the outer layer data of the building and the main building segments, a triangular mesh model is constructed to obtain the building complex hierarchical model. The building complex hierarchical model is a high-resolution original building complex model, which realizes the step-by-step fine representation from the low-resolution original building complex model to the high-resolution building complex hierarchical model. Therefore,

[0101] This solution realizes a method for constructing a hierarchical model of a building complex that can ensure the consistency of the model structure within the same LOD level without the need for pre-defining the levels.

[0102] In some embodiments, the process overview of the building complex level model construction method is as follows: Figure 3 shown.

[0103] In an exemplary embodiment, Figure 4 As shown, the building complex hierarchical model construction method further includes steps 402 to 408. Among them:

[0104] Step 402 : constructing a shape-fidelity model through the intersection of the obtained initial associated structure segment set and the obtained initial main structure segment set to obtain an initial shape-fidelity sub-model.

[0105] In some embodiments, the initial associated structure segment set and the initial main structure segment set may be obtained through point cloud data of a certain building directly input by a user, or may be obtained through a database, but is not limited thereto.

[0106] The expression of the initial shape-preserving sub-model is as follows:

[0107]

[0108] Among them, M is the initial main structure segment set, I is the initial associated structure segment set, and P(X) represents the voxel center point surrounded by the structure segment set X.

[0109] In some embodiments, the shape fidelity term of the initial shape-fidelity sub-model is defined by the intersection of the obtained initial associated structure segment set and the obtained initial main structure segment set. , shape fidelity term It is used to reflect the overall shape of the initial main structure segment set and the initial associated structure segment set, that is, The larger it is, the higher the fidelity of the initial set of principal structure segments.

[0110] Step 404 , constructing a simplicity model through the plane area ratio between the initial associated structure segment set and the initial main structure segment set to obtain an initial simplicity sub-model.

[0111] The expression of the initial simplicity sub-model is as follows:

[0112]

[0113] in Indicates the total plane area contained in the initial main structure segment set, represents the total area of ​​the planes contained in the initial set of associated segments, and The larger it is, the larger the area of ​​the main structure segment in the corresponding building is, which means that the simplicity of the current main structure segment is poor.

[0114] In some embodiments, the simplicity of the main building is further ensured by limiting the area of ​​the plane of the initial main structure segment set in the plane of the initial associated structure segment set. In some extreme cases, the initial simplicity sub-model is maximized. , which can simplify the geometric structure of the main building layer to the greatest extent.

[0115] Step 406 , constructing a semantic model through the cross similarity between the initial associated structure segment set and the initial main structure segment set to obtain an initial semantic sub-model.

[0116] In some embodiments, the process of calculating the cross-similarity between the initial set of associated structure segments and the initial set of main structure segments is as follows:

[0117] First, define two structure segments s and The measurement method between is as follows:

[0118]

[0119] in, is a function used to measure the distance between structural segments. The smaller the distance between two segments, the higher their similarity. The distance between structural segments is defined by two components: shape distance and scale distance. represents the process of making s isotropic by anisotropic scaling, Represents the D2 descriptor of x. If is defined as the main eigenvalue vector of the covariance matrix of s, then Defined as based on and Secondly, based on the similarity between segments, a cross analysis is performed on two structural segments s and S of different buildings, and the expression is:

[0120]

[0121] in, represent Assemble with another building's structural segment The distance between Representative based on After defining the cross-similarity, the consistency term of the initial semantic sub-model can be calculated It takes into account the mutual similarity between different segments and formulates a .

[0122] In some embodiments, the rule of the real scene can be that the structure segments outside the main structure segment usually include windows, fences, roof water tanks, air conditioning units and other components that are not designed separately, so the formulated Structural segments that match well with other segments are encouraged to be removed from the main structural segments, because structural segments that match well with other segments represent repeated structures. In practice, the expression of the initial semantic sub-model is as follows:

[0123]

[0124] in, is the initial main structure segment set of all buildings, The initial set of associated structural segments for all buildings.

[0125] Step 408 , constructing an energy model based on the shape-fidelity sub-model, the initial simplicity sub-model and the initial semantic sub-model to obtain an initial energy model.

[0126] In some embodiments, the weight ratios of the initial simplicity sub-model and the initial semantic sub-model are set in combination with specific application requirements, and then the buildings whose shape fidelity corresponding to the shape fidelity sub-model is greater than the fidelity threshold are numerically accumulated and summed up with the corresponding initial shape fidelity sub-model, initial simplicity sub-model and initial semantic sub-model, so as to construct an initial energy model.

[0127] In this embodiment, since the building complex hierarchical model constructed by this scheme has better fidelity at the same complexity, the rendering speed can be further improved, and higher scene rendering quality can be guaranteed at the same rendering efficiency as other technologies. At the same time, it is more suitable for downstream applications such as urban planning that have high requirements for consistency within the LOD level.

[0128] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0129] Based on the same inventive concept, the embodiment of the present application also provides a building complex hierarchical model construction device for implementing the above-mentioned building complex hierarchical model construction method. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above-mentioned method, so the specific limitations in one or more building complex hierarchical model construction device embodiments provided below can refer to the limitations of the building complex hierarchical model construction method above, and will not be repeated here.

[0130] In an exemplary embodiment, Figure 5 As shown, a building complex hierarchical model construction device is provided, comprising: a plane detection module 501, a structure segmentation generation module 502, a model optimization module 503, a main body division module 504, an outer level division module 505 and a model construction module 506, wherein:

[0131] The plane detection module 501 is used to perform plane detection on the original model data through the regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model;

[0132] The structure segmentation generating module 502 is used to perform voxel grouping on the initial plane data to obtain voxel plane groups, and perform plane aggregation on the voxel plane groups to obtain aggregated structure segments;

[0133] A model optimization module 503 is used to optimize the preset initial energy model according to the aggregate structure segmentation to obtain a target energy model;

[0134] The main body division module 504 is used to divide the aggregate structure segments into main body levels according to the target energy model to obtain the building main body segments and the building outer layer segments;

[0135] The outer layer division module 505 is used to divide the outer layer segments of the building into outer layers to obtain the outer layer data of the building;

[0136] The model building module 506 is used to construct a triangular mesh model according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

[0137] In some embodiments, the structure segmentation generation module 502 is further used to voxelize the original model data to obtain voxel units; and to plane group the initial plane data according to the voxel units to obtain voxel plane groups.

[0138] In some embodiments, the structure segment generation module 502 is further used to perform overlap evaluation on the voxel plane group and obtain the evaluation result; and aggregate the voxel plane group according to the evaluation result to obtain the aggregated structure segment.

[0139] In some embodiments, the model optimization module 503 is further used to perform binary linear programming on the aggregate structure segments to obtain linear programming results; and perform linear programming solution on the preset initial energy model according to the linear programming results to obtain the target energy model.

[0140] In some embodiments, the outer layer division module 505 is further used to perform spectral clustering analysis on the outer layer segments of the building to obtain outer layer clustering results; and perform hierarchical division on the outer layer segments of the building according to the outer layer clustering results to obtain outer layer data of the building.

[0141] In some embodiments, the device also includes: an energy model construction module, which is used to construct a shape-fidelity model through the intersection of the acquired initial associated structure segment set and the acquired initial main structure segment set to obtain an initial shape-fidelity sub-model; construct a simplicity model through the plane area ratio between the initial associated structure segment set and the initial main structure segment set to obtain an initial simplicity sub-model; construct a semantic model through the cross-similarity between the initial associated structure segment set and the initial main structure segment set to obtain an initial semantic sub-model; and construct an energy model based on the shape-fidelity sub-model, the initial simplicity sub-model and the initial semantic sub-model to obtain an initial energy model.

[0142] In the above-mentioned building complex hierarchical model construction device, the original model data is plane-detected by the regional growth model to obtain the initial plane data, wherein the original model data is used to store the original building complex model, and then the initial plane data is voxelized and grouped to obtain the voxel plane group, and the voxel plane group is plane-aggregated to obtain the aggregated structure segment, so that each building in the original building complex model is divided into multiple aggregated structure segments, which is more flexible than the pre-defined hierarchy method and also has better structural perception ability. Subsequently, the model is optimized according to the aggregated structure segment and the preset initial energy model to obtain the target energy model that can better perceive the detailed structure of the aggregated structure segment, and then the aggregated structure segment is divided into the main building segment and the outer building segment according to the target energy model. The hierarchical division through the optimized target energy model can ensure the structural consistency across buildings in the same level. Then, the outer layer segments of the building are divided into outer layers to obtain the outer layer data of the building. According to the outer layer data of the building and the main body segments of the building, a triangular mesh model is constructed to obtain a building complex hierarchical model. The building complex hierarchical model is a high-resolution original building complex model, which realizes the step-by-step fine representation from the low-resolution original building complex model to the high-resolution building complex hierarchical model. Therefore, this scheme realizes a building complex hierarchical model construction device that can ensure the consistency of the model structure within the same LOD level without pre-defining the level.

[0143] Each module in the above-mentioned building complex hierarchical model construction device can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0144] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 6As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store original model data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for constructing a hierarchical model of a building complex is implemented.

[0145] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0146] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0147] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0148] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0149] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0150] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0151] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0152] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for constructing a building complex hierarchical model, characterized in that: The method comprises: Performing plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model; voxelize and group the initial plane data to obtain a voxel plane group, and perform plane aggregation on the voxel plane group to obtain an aggregated structure segment; A shape-fidelity model is constructed by using the intersection of the obtained initial associated structure segment set and the obtained initial main structure segment set to obtain an initial shape-fidelity sub-model; A simplicity model is constructed by using the plane area ratio between the initial associated structure segment set and the initial main structure segment set to obtain an initial simplicity sub-model; Constructing a semantic model by cross-similarity between the initial associated structure segment set and the initial main structure segment set to obtain an initial semantic sub-model; Constructing an energy model according to the shape-fidelity sub-model, the initial simplicity sub-model and the initial semantic sub-model to obtain an initial energy model; Performing model optimization according to the aggregate structure segmentation and the initial energy model to obtain a target energy model; According to the target energy model, the aggregate structure segment is divided into main hierarchical groups to obtain building main segments and building outer segments; Dividing the outer layer of the building into segments into outer layers to obtain outer layer data of the building; A triangular mesh model is constructed according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

2. The method according to claim 1, characterized in that The voxelizing and grouping the initial plane data to obtain a voxel plane group includes: voxelize the original model data to obtain voxel units; The initial plane data is plane-grouped according to the voxel units to obtain a voxel plane group.

3. The method according to claim 1, characterized in that The performing planar aggregation on the voxel plane group to obtain aggregated structure segments comprises: performing overlap evaluation on the voxel plane group and evaluating the result; The voxel plane groups are aggregated according to the evaluation results to obtain aggregated structure segments.

4. The method according to claim 1, characterized in that The step of performing model optimization according to the aggregate structure segmentation and the initial energy model to obtain a target energy model includes: Performing binary linear programming on the aggregation structure segment by segment to obtain linear programming results; The initial energy model is solved by linear programming according to the linear programming result to obtain the target energy model.

5. The method according to claim 1, characterized in that The step of dividing the outer layer of the building into segments to obtain outer layer data of the building includes: Performing spectral clustering analysis on the outer layer segments of the building to obtain outer layer level clustering results; The outer layer segments of the building are hierarchically divided according to the outer layer clustering result to obtain the outer layer data of the building.

6. A building complex hierarchical model construction device, characterized in that: The device comprises: A plane detection module is used to perform plane detection on the original model data through a regional growth model to obtain initial plane data; wherein the original model data is used to store the original building complex model; A structure segmentation generation module, used for performing voxel grouping on the initial plane data to obtain a voxel plane group, and performing plane aggregation on the voxel plane group to obtain an aggregated structure segment; A model optimization module is used to construct a shape-fidelity model through the intersection of the obtained initial associated structure segment set and the obtained initial main structure segment set to obtain an initial shape-fidelity sub-model; construct a simplicity model through the plane area ratio between the initial associated structure segment set and the initial main structure segment set to obtain an initial simplicity sub-model; construct a semantic model through the cross-similarity between the initial associated structure segment set and the initial main structure segment set to obtain an initial semantic sub-model; construct an energy model according to the shape-fidelity sub-model, the initial simplicity sub-model and the initial semantic sub-model to obtain an initial energy model; and optimize the initial energy model according to the aggregated structure segmentation to obtain a target energy model; A main body division module, used for performing main body hierarchical division on the aggregate structure segment according to the target energy model to obtain a building main body segment and a building outer layer segment; An outer layer division module is used to divide the outer layer segments of the building into outer layers to obtain outer layer data of the building; The model building module is used to construct a triangular mesh model according to the building outer layer data and the building main body segmentation to obtain a building complex layer model; the building complex layer model is a high-resolution original building complex model.

7. The device according to claim 6, characterized in that The structure segmentation generation module is also used to voxelize the original model data to obtain voxel units; and to plane group the initial plane data according to the voxel units to obtain voxel plane groups.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images